In exercises that call for two-sample procedures, you may use either of the two approximations for the degrees of freedom that we have discussed: the value given by your software or the smaller of and . Be sure to state clearly which approximation you have used.
7.57 Drive-thru customer service.
QSRMagazine.com assessed 1855 drive-thru visits at quickservice restaurants.28 One benchmark assessed was customer service. Responses ranged from “Rude (1)” to “Very Friendly (5).” The following table breaks down the responses according to two of the chains studied.
drvthru
Rating | |||||
---|---|---|---|---|---|
Chain | 1 | 2 | 3 | 4 | 5 |
Taco Bell | 0 | 5 | 41 | 143 | 119 |
McDonald’s | 1 | 22 | 55 | 139 | 100 |
7.57
(a) The problem with averages on rating is that there is no guarantee the differences between ratings are equal, so that going from a rating of 1 to 2, and 2 to 3, etc., are equal. Taking averages assumes this so it is likely not appropriate. (b) The data are ratings from 1–5; as such they certainly will not be Normally distributed but because and outliers are not possible, the procedures can be used. (c) McDonald’s: . Taco Bell: . (d) . The data are significant at the 5% level, and there is evidence the average customer ratings between the two chains is different.